KO Knockout Blackjack - some clarification with 4 decks

[bowlofrice]

Hi,

I can count comfortably with KO and apply the preferred matrix as needed.

I also apply, some weighting to decisions based on the number of decks remaining as has been discussed on other threads i.e. KO can undervalue the true count very early and can over value later on etc.

I want to incorporate some of the full matrix plays i.e. hit 12 vs 4 at -7 or lower and double 9 vs 2 at -4 (page 164).

I play in 4 deck games quite a bit and although most of the matrices for 6 decks apply to 4 decks there is something that got me thinking.

Double 9 vs 2 @ -4 in a 6 deck game: -4 is the key count.

However on page 171 in a 4 deck game it says the key count in a 4 deck game is at -1.

Does this mean the 9 vs 2 @ -4 in a 4 deck game is inaccurate?

Also I can't see how to derive the key count from the IRC based on the number of decks

i.e

(number of decks * 4) + 4 is formula for the IRC - straightforward - but where was the key count derived from the IRC?

 

[Taff]

Double 9 vs 2 @ -4 in a 6 deck game: -4 is the key count.

However on page 171 in a 4 deck game it says the key count in a 4 deck game is at -1

You're confusing 2 seperate things. The -4 in the six deck game relates to the point at which you depart from basic strategy in order to make that play. This is not the key count. The -1 in the four deck game is the key count which is always 11 points above the IRC.

Does this mean the 9 vs 2 @ -4 in a 4 deck game is inaccurate?

Yes. I dont have the book in front of me but I believe there is no matrix table for a 4 deck game. If you true count K.O you can easily customize your own deviation points and cluster them under +1, +2 etc rather than trying to remember points along the running count to deviate. Easier in my opinion.

Also I can't see how to derive the key count from the IRC based on the number of decks

Are you asking how to true count K.O.??

 

[bowlofrice]

Thanks for your reply.

Are you asking how to true count K.O.??

No, sorry, probably was not being clear. Definitely not trying to do this.

I appreciate KO never be as accurate as Hi-Lo and the differences are explained in the book but once the pivot point has been passed I have seen enough positive runs around this count value to prove to me that is effective assuming you don't get screwed by variances. I have also seen plenty of situations where the count is >+4 and variance destoys you, but enough positive runs to warrant its contuined use over learning Hi-Lo for the moment.

Given this, instead of doing a poor job of learning Hi-Lo which will take time I want to try and squeeze every last drop of value out of an accurate KO count.

Since Wonging in/out is not practical I will have to sit through shoes at minimum bet while the count is poor and as such I was looking at putting in the effort of learning the Full KO Matrix.

The full table is in the book for 6 decks but not for 4 but it does say some of their tables for 6 decks apply to 4 but I am not sure if that applies for the full matrix which has these plays like 12 vs 4 - hit while count < =7 etc.

They also have some plays which seem to supercede the preferred matrix by a significant amount.

e.g.

16 vs 10 = stand at >= -4 (preferred matrix)
16 vs 10 = stand at >= -8 (full matrix)

and this one:

16 vs 9 = stand at 4 (preferred matrix)
16 vs 9 = stand at >=10 (full matrix) <- this is quite a big difference from +4 - this is why I want to understand how theses are achieve. Not taking a card with hard 16 vs 9 at either positive counts of 4 or 10 respectively.

I am concerned if apply the six deck matrix full for my 4 shoe games there are some aspects where I will be fooling myself!

 

[Taff]

I want to try and squeeze every last drop of value out of an accurate KO count.

To be honest the index plays you're focusing on aren't going to do this for you. 16v10,12v4 and 9v2 aint worth a tin of beans, the first 2 being almost a coin toss and the third as a low index play of TC1. Many of the other full Matrix plays can simply be played at the pivot point. 15v10,9v7,12v2,8v5/6 without losing any noticable accuracy. Daniel Davrot does this in Colour of Blackjack. I would never split 10's but that's a personal thing.

 

[Taff]

16 vs 9 = stand at 4 (preferred matrix)
16 vs 9 = stand at >=10 (full matrix)

I have no idea why the disparity here. It's too big. If you look one line down 15v10 it is played at 5. As they are almost, but not quiet identical plays I'm lost here. As Ko preferred has both plays made at the pivot point you would be best to go with that IMHO.

 

[bowlofrice]

I have no idea why the disparity here. It's too big. If you look one line down 15v10 it is played at 5. As they are almost, but not quiet identical plays I'm lost here. As Ko preferred has both plays made at the pivot point you would be best to go with that IMHO.

I am glad you thought this was a bit odd. I was coming to the conclusion if the shoes is rich enough to stand at 4 then how does the full or more comprehensive matrix have you keep drawing all the way up to 10! Maybe it's a typo.

For reference and other people reading this thread I have attached the relevant pages:

Appendix VII: The K-O for 4 Decks
Following the publication of the first edition of KnockOut Blackjack, we received several inquiries for information about applying K-O in the 4-deck game. It is included here. The key count for 4 decks is +11 above the IRC. The pivot point is +16 above the IRC. Take insurance, as always, at the count which is one below the pivot point. Therefore, based on the 4-deck game's standard IRC of12, the key count is -1, the pivot point is +4, and insurance is taken at +3. As in 6- and 8-deck shoes, omit Category C plays altogether in the strategy matrix. There are no changes in late surrender. The expectation for the K-O Preferred for 4 decks (517, DOA, DA5, 75% pen.) with a 1-6 spread is .75%. The expectation with a 1-8 spread is .95%.

Many of the other full Matrix plays can simply be played at the pivot point. 15v10,9v7,12v2,8v5/6 without losing any noticable accuracy

So using preferred as a baseline I was thinking of adding in from the full things like

12 vs 2 @ 2 instead of 4
12 vs 3 @ -2 instead of 4
12 vs 4 @ -7 instead of always stand
13 vs 2 @ -13 or less instead of always stand

Double against As etc. don't factor in for me because I'm in Europe and there is no peek.

I appreciate there are some plays in the full that are so rare due to the high count required:

16 vs 8 @ 16
16 vs 9 @ 10

This is a strange one:

Hard 16 vs 10 @ -8 What's the deviation? It's a hit under basic, it's hard so not a split and at -8 there is still a chance of small cards.

 

[Taff]

Hard 16 vs 10 @ -8 What's the deviation? It's a hit under basic, it's hard so not a split and at -8 there is still a chance of small cards

-8 in the six deck game is the point at which you have the advantage derived from an equal number of high and low cards left in the shoe. The pivot point. Standing 16 v 10 @ the pivot point and above. In the 4 deck game it's -1.

12 vs 2 @ 2 instead of 4
12 vs 3 @ -2 instead of 4
12 vs 4 @ -7 instead of always stand
13 vs 2 @ -13 or less instead of always stand

These look about right. Please understand I play a very customized version of K.O with a true count method and my own indices. Maybe someone who plays K.O pref can chip in and give better guidance than I. I'm looking at your index plays from a deck adjusted point of view.

 

[bowlofrice]

-8 in the six deck game is the point at which you have the advantage derived from an equal number of high and low cards left in the shoe. The pivot point. Standing 16 v 10 @ the pivot point and above. In the 4 deck game it's -1.

I thought the pivot point is +4 when both KO and Hi Lo converge and -8 is a long way off that.

From my understanding in the book the Key Count (which you may have been referring to?) where the equal number of cards occurs I thought was at -4 for 6 decks (from page 86 attached) and then they made the clarification that for 4 decks it was at -1?

 

[Taff]

Sorry. You're right I'm wrong. It's because I set the pivot point at 0 not +4 so I was referring to my customized version

 

[bowlofrice]

OK, so in standard KO IRC what they are saying is start standing on 16 vs 10 at -8 instead of -4. That's quite a big difference, most shoes float around 0 to - 12 once they get going.

 

[Taff]

I just re read your initial question and realised why I may have answered it wrong based on my own version of K.O. yes you would double 9v2 at -4 in the six deck as this is YOUR key count. -1 in the 4 deck. I dont have a key count at all rather fixed points at which I raise my bet according to how many decks are left undealt. It really is quite simple if you want to get the most out of K.O.

 

[bowlofrice]

I dont have a key count at all rather fixed points at which I raise my bet according to how many decks are left undealt. It really is quite simple if you want to get the most out of K.O.

Ok. I am aiming to follow the full matrix instead of preferred as a way of getting a little bit more edge during poor counts. Regarding the number of decks left I increase/decrease bet ammounts based of a decks left but only in a very abstract way. There is another thread on here somewhere with an excel spreadsheet attached that shows that KO underreads at the start of the shoe and can overread towards the end. To that end is the IRC is -20 and after only 1/4 to 1/2 deck dealt the RC is around -10 I will increase by bet and towards the very end of the shoe is we are on the pivot I will be slightly more cautious with sizing. Since I was moving to full matrix and 4 deck games I wanted to know if I saw -4 in the full matrx for the published 6 deck game matrix that would be comparable to -4 in the deck game since 4 and 6 deck stategy seems to be mostly interchangable when doing preferred.

 

[gronbog]

To be honest the index plays you're focusing on aren't going to do this for you. 16v10,12v4 and 9v2 aint worth a tin of beans

These are all part of the I18.

 

[Taff]

To that end is the IRC is -20 and after only 1/4 to 1/2 deck dealt the RC is around -10 I will increase by be

Regarding the number of decks left I increase/decrease bet ammounts based of a decks left but only in a very abstract way.

Why do it in an abstract way.?? You're nearly there. Actually with 5 decks left to play the trigger point to raise bet is -11. It then goes up in 3's according to decks left to be dealt. So:
4 decks trigger @-8
3 decks trigger @ -5
2 decks trigger @-2
1 deck trigger @+1

Right. Now all you have to do is create your own bet ramp between the trigger point and the pivot point. You can see from the above how you miss betting opportunities early on in the shoe with the trigger point way below the key count and, conversely overbet towards the end with the trigger point way above it.

 

[Taff]

These are all part of the I18.

Yes. But poster was talking about using them in neutral counts.

 

[gronbog]

You should use them when called for, neutral count or otherwise. They are still among the top 18 indices. Period.

 

[Taff]

Yes I know the I18. So when is 9v2 called for in a neutral count.? Am I missing something.? The I18 states Tc1. So at Tc 0 are you saying we should also use it.??

 

[gronbog]

Depends on your definition of a neutral count i guess. 0 is a negative EV count. 1 is a positive EV count in many systems. My response was directed at your statement that 9vs2, 16vs10 and 12vs4 "ain't worth a tin of beans".

These plays should be made at the proper indices for the system in use and are valuable when used in that way. No need to make it more complicated than that.

 

[Taff]

These plays should be made at the proper indices for the system in use and are valuable when used in that way.

Which was exactly the point I was making, in particular with regards to 9v2 which the poster was deliberating over using at a neutral count. I had no intention of suggesting these plays are worthless just that they are so at the incorrect points in the count. This play is triggered at the key count in K.O where under normal conditions you barely have an edge. As with most things 'blackjack' the system was developed over the pond with that game in mind. Poster is playing ENHC so at the key count he has no edge whatsoever. Playing 9v2 AT THIS POINT is worthless. I will concede that my comments, taken in isolation, look like I'm dismissing the value of these index plays. Believe me I'm not.

 

[bowlofrice]

Why do it in an abstract way.?? You're nearly there. Actually with 5 decks left to play the trigger point to raise bet is -11. It then goes up in 3's according to decks left to be dealt. So:
4 decks trigger @-8
3 decks trigger @ -5
2 decks trigger @-2
1 deck trigger @+1

Right. Now all you have to do is create your own bet ramp between the trigger point and the pivot point. You can see from the above how you miss betting opportunities early on in the shoe with the trigger point way below the key count and, conversely overbet towards the end with the trigger point way above it.

So my eyeball was close. Good so I can use these (just to be clear this is using the standard IRC number and not your offset one!?)

Going back to the original question in relation to 4 decks here I have transcribed all the full matrix plays that I believe are availble to me and their associated count - this is from page 164 - 6 decks

<=8 vs 4
+9 Hit / Double

<=8 vs 5
+5 Hit / Double

<=8 vs 6
-1 Hit / Double

9 vs 2
-4 Hit / Double

9 vs 7
+3 Hit / Double

12 vs 2
+2 Hit / Stand

12 vs 3
-2 Hit / Stand

12 vs 4
-7 Stand / Hit

14 vs 10
+13 Stand / Hit

15 vs 10
+5 Hit / Stand

16 vs 8
+16 Hit / Stand

16 vs 9
+10 Hit / Stand

16 vs 10
-8 Hit / Stand

16 vs A
+15 Hit / Stand

Can't use the following due to risk of BJ in ENHC
10 vs 10
+3 Hit / Double

10 vs A
+4 Hit / Double

11 vs A
-2 Hit / Double

My concern is that on page 163 there is full matrix for double deck and has many more index plays which is to be expectd - now double deck is obviously widly different from 6 deck but 4 decks is in the middle. Despite the authors stating 4/6 stats are interchangable with their brief addendum I am concerend that some of these 6 decks index plays are off for 4 decks. One large difference from preferred to full is:

Preferred:
16 vs 10
-4 Hit / Stand

Full:
16 vs 10
-8 Hit / Stand

This might be quite bad at 4 decks. I just don't know or understand how to exptrapolate / interpolate etc.